Functional magnetic resonance imaging (fMRI) relies on the coupling of blood flow changes to neuronal activity. These blood flow changes in turn alter the concentration of oxygen in the local blood pool surrounding the neurons. Because the susceptibility of oxyhemoglobin and deoxyhemoglobin are different, it is possible to make the MR signal sensitive to oxygen concentration; this is termed blood oxygenation level dependent (BOLD) signal. The BOLD signal change from the rest condition to the active condition is approximately 1-3%. Detection of the BOLD signal requires a robust paradigm, preprocessing of the signal, and statistical modeling to generate activation maps. All of this is necessary because of the weak BOLD contrast that exists in hemodynamically normal subjects. When a subject has compromised vasculature and impaired vasoreactivity from a carotid stenosis or occlusion, we have demonstrated that the BOLD signal response is absent or severely altered beyond normal detection. This is a very important issue as fMRI moves into the clinical setting and is used to study stroke recovery or aging populations. For example, in stroke recovery one may detect activation increases over time and assume they are due to neuronal recruitment or rehabilitation; however, the activations may be the result of developing collateral flow or recovery of vasoreactivity. One of the goals of this proposal is to investigate and characterize the BOLD response in the face of altered hemodynamics. A limitation to the field of functional MRI is the small difference in BOLD signal used to detect activations. We have shown that the physiologic interaction of caffeine signigicantly increases the BOLD contrast (-40% at 1,5T and 80-170% at 3T). In this proposal, the mechanisms of caffeine will be explored in normal subjects using fMRI, MR based perfusion and transcranial Doppler ultrasound to characterize the BOLD signal and the cerebral blood flow response. Individual components will be invesitgated in order to create a simple model of the interations with BOLD contrast. The improvement in BOLD contrast can be used to improve the temporal and spatial resolution or to allow the investigation of more subtle cognitive paradigms.The theme of this proposal is to investigate the impact of physiologic mechanisms (reactivity, flow, hematocrit, neuronal activity) on the generation of BOLD signal and more importantly BOLD contrast.